Management of iris melanoma with secondary glaucoma.

Abstract

CASE REPORT A 53-year-old patient with a history of an anomalous disc and amblyopia in the left eye was diagnosed with an iris melanoma in the right eye, which was discovered during a routine eye examination. The lesion extended into the drainage angle, obliterating the angle for greater than 9 clock hours, and had grown onto the inferior corneal endothelium. The melanoma extended posteriorly over the pars plicata, as viewed by ultrasound. Visual acuity on presentation was 20/20 in the right eye, and 20/200 in the left. Intraocular pressure was 35 OD and 20 OS. Ophthalmoscopy revealed possible early glaucomatous changes to the right optic disc and an anomalous disc in the left eye. Visual fields were full to SITA perimetry. After retinal and oncological consultation, and discussion with the patient, it was decided to follow the tumor as the risk of metastasis was low, the lesion was too large for complete resection, and radiation therapy was inappropriate due to the anterior location of the tumor. The patient was treated medically for his elevated intraocular pressure, and baseline stereoscopic disc photography and optic disc imaging studies were performed. Over the following 8 months, there was no interval change in tumor size; however, the intraocular pressure was only lowered to the upper twenties on topical carbonic anhydrase inhibitor, β blocker and latanoprost. Clear progression of optic disc changes and visual field progression was evident on follow-up evaluation. Questions: What is the incidence of metastasis of an iris melanoma that invades the drainage angle? How would you evaluate this patient? How would you treat this patient with regards to his tumor? This patient was seen 5 years before with an iris nevus in the area. What high risk characteristics could possibly indicate malignant transformation? How would you manage this patient with a suspicious lesion? Given the failure of medical therapy and progression of glaucoma, how would you seek to control his progressive disease? What are the potential risks of each possible treatment option? Describe your subsequent interventions should your initial treatment fail. COMMENTS Comments byIvan Goldberg, FRACS,Sidney Eye Hospital, Sydney, Australia. A metastasis rate of around 3% is generally quoted for iris melanomas. However, if the drainage angle is involved, a doubling of this rate is often stated. A careful patient evaluation with detailed clinical examination and appropriate imaging techniques, together with retinal and oncological consultations is essential. A suspicious lesion needs careful documentation of size, position, color, and texture, followed by observation. Photography and ultrasonography are vital aids. Little or no change over time would suggest a more benign spindle cell melanoma, compared with the more usual aggression of an epithelioid tumor. More rapid expansion and/or extension would call for local excision, plaque application or proton beam treatment, depending on size and location. Enucleation remains the definitive treatment for rapid growth. With respect to tumor management, initial observation seems appropriate. In this case, no growth was noted for 8 months. Should there be signs of extension, one needs to consider enucleation versus local therapy. The implication from this case report is that the right eye, with the tumor, is the only eye. However, there are levels of amblyopia. How profound is the visual loss in the left eye? How effective is the left peripheral visual field? How easily could the patient manage solely with the left eye? If life itself is at stake, and enucleation of the right eye will enhance the chances of survival, would the patient be (relatively) content with the vision afforded by the left eye? If yes, then such surgery may be indicated in the right. If the patient really is uniocular, and local therapy is sought, the 9 o'clock hour extent and anterior location of the melanoma make treatment challenging. Any plaque would need to be far too large, local excision and/or proton beam therapy, too difficult. A combined approach may work—partial excision and plaque application to the remainder. Oncological advice is crucial. High risk characteristics of a pigmented lesion indicative of more likely malignant behavior or transformation include overall tumor size at diagnosis, observed growth and the rate of the growth, posterior segment extension, the presence of secondary glaucoma, and/or spontaneous hyphema. In this case, medical therapy has not necessarily failed. There are two options remaining to be explored. An α-2 agonist such as brimonidine may well prove helpful, at least for a time, and a systemic carbonic anhydrase inhibitor may be more effective in lowering intraocular pressure (IOP) than the topical preparation. Any time gained with IOP control is valuable. Should progressive damage occur despite these measures, or should IOP remain in the high twenties, the options are either to increase aqueous outflow or to reduce inflow. Laser trabeculoplasty is contraindicated, not only because of the danger of showering tumor cells within the eye, but also because of the extreme unlikelihood of achieving any IOP reduction. Drainage surgery carries the risk of facilitating tumor spread into the extraocular and orbital tissues. A seton, such as a Molteno, Baerveldt, or Ahmed implant also increases the risk of extraocular spread. The safer alternative is to reduce inflow. Ciliary body destruction can be achieved by freezing or laser applications ab externo or ab interno. Because cyclocryotherapy causes more pain and inflammation than laser techniques, it can be discounted. Ab interno laser involves significant globe invasion, which is also inadvisable. The technique most likely to succeed with the least chance of producing problems is ab externo Nd-YAG or diode laser therapy. Contact applications can be honed onto the ciliary epithelium more accurately than noncontact approaches. I recommend laser applications to two quadrants (170 degrees) of the ciliary body, avoiding the 3 and 9 o'clock positions to minimize the risks of anterior segment ischemia. Usually three or four applications are needed per quadrant, and the laser power is adjusted just to fail to produce a barely audible `popping' sound. A week is allowed to assess the response, and if insufficient IOP reduction is achieved, another two quadrants are treated, one being retreated, and the second being treated de-novo. A third treatment of two quadrants, again retreating one quadrant, is also possible, if necessary. By gradually increasing the extent of ciliary body treatment, one attempts to minimize the risks of overtreatment, which may result in hypotony or even phthisis. Such a scaled program attempts to titrate inflow reduction against required IOP reduction. Comments bySteven L. Mansberger, MD,Devers Eye Institute, Portland, Oregon. Management of this patient contains several levels of complexities. With any difficult management decisions, it is important to identify the goals of treatment and, of course, to `do no harm'. The extent of the tumor can be evaluated with clinical examination, standard photography, and advanced imaging technologies. External examination of the sclera with a penlight can determine if the tumor has an extrascleral extension; transillumination of the eye can identify the margins of the lesion; B-scan ultrasound may delineate its posterior dimensions; and a high frequency ultrasound biomicroscope (UBM) may be used to determine its anterior extent and rule out egress through the trabecular meshwork and emissary channels. Photography of the iris and angle should be carried out periodically to document any change in the size, shape, or appearance of the tumor. If the iris melanoma were noted five years previously to be an iris nevus, this would indicate that the rate of growth of the tumor is rapid. Other signs of an aggressive tumor would be the presence of sentinel vessels in the region of the tumor and increased vascularity with indocyanine green (ICG) or fluorescein angiography. For treatment of a tumor without extrascleral extension or metastasis, I would confer with a colleague experienced with uveal melanomas. The second management decision is treatment of the elevated IOP and glaucomatous progression. Sometimes, when a patient has unilateral glaucoma, the IOP from the contralateral eye may be used as a guide to determine a target IOP for the glaucomatous eye. I would consider switching him from the topical carbonic anhydrase inhibitor and the β blocker to Cosopt, which would decrease the number of drops administered and may help with short term and long term compliance. The patient may also be started on an α-2 agonist, such as brimonidine tartrate, which may give additional IOP lowering. Latanoprost may theoretically increase the risk of metastasis as it reduces hydrostatic resistance to outflow through the uveal tissue. If there were evidence of a therapeutic effect from latanoprost, I would continue to prescribe it after the patient has been informed of this theoretic risk. After treatment of the tumor, argon laser trabeculoplasty may be attempted over the 90 degrees of meshwork that is uninvolved by the tumor extension. Hypothetically, the success may be increased due to the increased pigment in the angle, but decreased due to only 90 degrees of the angle being available for treatment. A so-called nonpenetrating procedure, such as a viscocanalostomy, would probably not be beneficial, because the location to aqueous resistance to outflow would not be bypassed by this procedure. In addition, there is risk of extrascleral spread if microperforations occur during the procedure. Comments byJerry A. Shields, MD and Carol L. Shields, MD,Wills Eye Hospital, Philadelphia, Pennsylvania The management of a patient with iris melanoma with secondary glaucoma can be challenging. The ultimate decision regarding management of such a patient must take into account several factors. These include the risk of metastatic disease, size and growth pattern of the tumor, patient age and general health, status of the opposite eye, visual potential in the affected eye, and the response to glaucoma therapy. The patient described here is a healthy 53-year-old and the melanoma involves the angle for 9 clock hours. The visual acuity is good but there are early glaucomatous changes in the optic disc and visual fields. There is difficulty controlling the glaucoma medically. The opposite eye has amblyopia with a visual acuity of 20/200. There is sufficient information on diffuse iris melanomas to provide general answers to the questions posed. Concerning the possibility of metastasis, iris melanoma carries a 3 to 6% metastatic rate at 5 years after treatment. Recently identified, statistically significant risk factors for metastasis from iris melanoma include increasing patient age, elevated intraocular pressure, angle involvement by tumor, extraocular extension, and prior surgical treatment before referral to an ocular oncology center. 1 Iris root or angle involvement, as seen in this patient, imparts a slightly greater chance of metastasis. The patient under discussion has a diffuse iris melanoma that is too large for successful removal by iridocyclectomy. Such patients were managed in the past by enucleation of the affected eye. Since the tumor is located in the patient's better eye, one would like to have a satisfactory alternative to enucleation. In this case the author states that radiation was deemed inappropriate due to the anterior location of the tumor. On the contrary, it has been reported that custom designed plaque irradiation is an acceptable option to enucleation for many nonresectable iris melanomas. 2 We reviewed our results of 14 iris melanomas treated with custom designed radioactive plaques and concluded that this was an effective method of controlling such tumors. 2 Observations on subsequent cases continue to support these results. We believe that brachytherapy, using the reported technique, would be the most acceptable option in the patient under consideration here, if there is evidence of tumor growth. 2 There is no evidence that the method of treatment of iris melanoma has any impact on patient survival. 1 The other question in this case relates to the management of the elevated IOP. It is unlikely that irradiation for the melanoma will have any beneficial effect on the pressure. If that is the case, then maximal medical treatment should be continued after plaque radiotherapy. If there is further damage to the optic disc and visual field, one should first consider argon laser trabeculoplasty or transscleral cyclophotocoagulation to an area not involved by the tumor. Open filtration should generally be avoided as long as possible since it could allow entrance of melanoma cells into the filtration bleb and possibly increase the chance of metastasis. If the patient eventually develops extremely poor vision and the pressure cannot be controlled, then enucleation would be justified. DISCUSSION Christopher A. Girkin, MD This case of an extensive iris melanoma with associated uncontrolled progressive glaucoma presents a fortunately uncommon treatment dilemma. The 5-year metastasis rate for iris melanoma is in the order of 3 to 6%. In this particular patient, the risk of metastasis is elevated due to the angle involvement and the raised intraocular pressure. However, the literature does not provide any clear evidence that the type of treatment has any effect on patient survival. The current strategies in dealing with iris melanomas have been summarized by Shields et al. 1 in their recent review of 169 consecutive cases. The strategy outlined by these authors involves a treatment plan aimed at treating the lesion with plaque radiotherapy. However, if the tumor had been confined to the central iris, was well circumscribed, and had not involved the angle, a surgical iridectomy could be performed. For tumors that involve the iris root, without seeding, a partial lamellar iridocyclectomy should be performed with a more posterior scleral flap and wider tumor margins. Any such of surgery should be performed by a specialist ophthalmic oncologist, with every attempt made to avoid seeding of the tumor. When diffuse seeding of the angle has occurred, which is the case in this patient, the treatment options in the event of documented growth are either enucleation or radiotherapy. This strategy was primarily applied to tumors with documented growth. We have followed our patient in conjunction with oncological consultation over the past year with anterior segment and angle photography, along with serial ultrasound examinations. There has been no indication of tumor growth and, after detailed discussion with the patient and continued oncological consultation, we have elected to continue to carefully follow this monocular patient. In counseling this patient, it was important to stress that the size of the lesion, its extent of angle involvement, and the elevation of the intraocular pressure all increase the 5-year risk of metastasis over the 3 to 6% supplied by the literature. Should signs of tumor growth become evident, anterior segment plaque radiotherapy would be the preferable treatment in order to maximize the chance of preserving useful visual function. With regards to the management of the patient's secondary glaucoma, we initially lowered the intraocular pressure from the middle thirties to the upper twenties using a three-drug regimen of topical carbonic anhydrase inhibitor, timolol, and latanoprost. Clear progression was evident by both stereo disc photography and visual fields. Subsequently, we have added brimonidine to his treatment regimen, and at the most recent follow-up examinations, IOP has been maintained at 20 mm Hg at 8am and 18 mm Hg at 3pm. Despite the difficulties of this four-drug regime, the patient is extremely motivated with good compliance. If IOP begins to rise or progression occurs (requiring a lower pressure level), other options would be to alter the medical therapy, such as the addition of pilocarpine, switching to an oral carbonic anhydrase inhibitor, or perhaps a trial of bimatoprost (however, its additive effect to latanoprost is unclear). One discussant mentions a potential for increasing the risk of metastasis with latanoprost due to increased uveoscleral clearance. The proposed mechanism of action of this drug is decreased interstitial resistance in the uveoscleral pathway presumably due in part to upregulation of matrix metalloproteinases (MMPs). Certainly, MMPs are found in the sclera as well, thus it might be possible for latanoprost to affect the scleral collagen. Although this might improve transscleral fluid diffusion, there is no evidence that cellular material could pass through this route as well. If this maximal medical therapy fails and surgical intervention becomes necessary, the options become somewhat limited. Incisional surgeries (i.e., trabeculectomy, shunt procedures, or viscocanalostomy) are clearly contraindicated due to the risk of extraocular spread. Laser trabeculoplasty to the remaining angle would be unlikely to result in significant IOP reduction, and carries with it, at least in theory, the risk of enhancing tumor spread. The safest surgical alternative would be a ciliary body destructive procedure. We prefer the transscleral diode laser, because it involves less inflammation and pain than cryotherapy. We preferably would begin treatment 180 degrees away from the tumor's location in the angle and, in this patient, initially treat the superior 180 degrees.